Salus.Coop Informe Final

SALUS.COOP
Towards citizen governance and management of health data
Ideas for change + Mobile World Capital Barcelona Foundation
December, 2016

This document was made by Ideas for Change with the
support of Mobile World Capital Barcelona Foundation.
Barcelona, December 2016

3
SALUS.COOP
IDEASFORCHANGE+MOBILEWORLDCAPITALBARCELONAFOUNDATION
Summary
Changes in the capacities and demands of citizens,
data-driven innovations and technological developments
in research invite us to reconsider the role of citizens in
the healthcare sector, both at individual and collective
levels. Mobile World Capital Barcelona Foundation and
Ideas for Change have investigated how such roles can
be reconfigured to accelerate research and innovation in
healthcare, and thus deliver positive social impact.
The goal of the collaboration agreement was
to explore the viability of a citizen-driven model of
collaborative governance and management of health
data, by proposing an approach that builds on three key
frameworks:
• Governance and relational framework
• Technological framework
• Legal framework.
To meet these goals Ideas for Change conducted
field and desk research. First, the state of the art of
existing health data sharing initiatives was reviewed and
interviews with experts and practitioners were performed.
These included health sector professionals, patient
associations, researchers and experts in the field of data
intensive technologies, among others.
Second, the collected data was analyzed by using
thematic analysis and key emergent themes were crafted.
These themes were then further analyzed and discussed
with a selected group of specialists in two validation
sessions.
Third, the research outputs were synthesized by
assembling higher-level findings that inform on a number
of recommendations to frame and activate the vision for
“Salus”. This report presents the outputs organized in three
chapters:
Chapter 1 describes the investigation. It discusses
why this study is timely and relevant, presents the vision
and objectives of the project, as well as the methodology
applied.
Chapter 2 presents the main findings and derives
four key pillars to structure a citizen-led governance
model for health data:
• Conditional donation;
• Collective benefits;
• Motivational incentives;
• Management of rights.
Chapter 3 introduces the proposed model. It
identifies the key agents and describes the different flows
of exchange that sustain the system of relationships
among them (economic, data and services). Furthermore,
it also explores the legal feasibility of the approach and
the existing technologies that could enable the model.
Finally, the report concludes with key findings
derived from this research.
EXECUTIVE SUMMARY

4
SALUS.COOP
TEAM
Javier Creus
CEO & Director of Strategy
Joan Guanyabens
Health expert
Mara Balestrini
Director of Research
Valeria Righi
Research
Andrea Barbiero
Health expert
Gabriela Masfarré
Strategy
Georgina Campins
Strategy
Daniela Arguedas
Creativity
Team

5
SALUS.COOP
TABLE OF CONTENTS
chapter 1:
This study ...6
1.1 Context ...7
1.2 Vision and objectives ...10
1.3 Methods ...12
chapter 2:
Results ...15
2.1 Mapping existing health data sharing
initiatives ...16
2.2 Results of interviews ...19
2.2.1 Universal benefits ...19
2.2.2 Terms for data donation ...20
2.2.3 Barriers ...22
2.3 Results of validation sessions ...24
2.4 Conclusion: the pillars for a citizen-
driven governance of health data ...27
chapter 3:
SALUS: towards citizen
governance and management
of health data ...30
3.1 Governance and relational framework ...31
3.1.1. Ecosystem of parties ...32
3.1.2. Data flow ...35
3.1.3. Service flow ...36
3.1.4. Economic flow ...37
3.1.5. Relational agreements ...39
3.2 Technological framework ...42
3.3 Legal framework ...46
chapter 4:
Conclusion ...49

7
SALUS.COOP
CHAPTER 1
THIS STUDY
1.1 Context
Advances in the field of Information and
Communication Technologies are generating enormous
amounts of data, which are laying the groundwork for
fostering collective intelligence and encouraging advances
in a range of different sectors.
Health is one of the sectors that produces the largest
volume of data, related to citizens’ health and lifestyle.
Within this sector, data are generated from different
sources, such as Electronic Health Records systems
(EHR), Personal Health Records systems (PHR), Hospital
Information Systems (HIS), Laboratory Information
Systems (LIS) and Radiology Information Systems (RIS or
PACS). New forms of data collection are being added to
these sources on an individual level (devices, sensors, etc.),
as well as on a social (social networks, blogs, etc.) and
environmental level (geo-positioning sensors).
The future of medical research will be significantly
reliant upon the potential for combining and integrating
all of these data sources1
. The benefits of using data in
research and medical services provision are tangible and
significant, as indicated by several studies2,3
.
Currently, the large amount and variety of data
1 Feldman, B., Martin, E. M., &Skotnes, T. (2012). Big
Data in Healthcare Hype and Hope. October 2012. Dr. Bonnie, 360.
2 Association of Medical ResearchCharities (2016) “A matter
of life and death: howyourhealthinformation can make a difference”
AMCR 2016.
3 TheBenefits of Data Sharing. In Olson, S., &Downey, A.
S. (Eds.). Sharingclinicalresearch data: workshopsummary. 2013
NationalAcademiesPress.
includes both structured data (e.g. analytical tests) and
unstructured data (e.g. images, videos and free text).
Unstructured data management has become one of the
main challenges faced by health administrations today. It
is estimated that 80% of health data are unstructured4
.
The lack of data structuring hinders the processes to
integrate and share that information. As a result, most
health data is currently stored in silos, which makes it
difficult to reuse, compare, and share.
However, there are several processes that can
transform unstructured data into accessible and reusable
data (e.g. capture, interoperability, and analysis processes).
One such example is the growing implementation
of big data processing systems. Within public health
systems, there are currently several projects aimed at the
integration and interoperability of health information on a
European level (e.g. Epsos5
), national level (e.g. HCDSNS6
)
and regional level (e.g. HC37
). These projects have digital
sites for citizens that allow them to check their personal
health information (e.g. La Meva Salut8
). Nevertheless,
none of these options legitimizes data ownership for
citizens, because, for instance, they do not provide
citizens with the tools to use, transfer and share their
4 Unstructured Data in ElectronicHealth Record (EHR)
Systems: Challenges and Solutions. ©2013 DATAMARK, Inc.
5 http://www.epsos.eu
6 https://www.msssi.gob.es/profesionales/hcdsns/home.htm
7 http://ticsalut.gencat.cat/ca/projectes_estrategics/
historia_clinica_compartida_a_catalunya/
8 https://lamevasalut.gencat.cat

8
SALUS.COOP
CHAPTER 1
THIS STUDY
information.
The Spanish Article 18.1 of Law 41/2002 on patient
autonomy, states that “The patient has the right to access
[...] the documentation of his or her medical records,
and the right to obtain a copy of the data contained
therein.” Accordingly, it can be claimed that data
ownership corresponds to citizens. However, currently,
citizens’ access to data is limited, since they can only
consult certain health information (e.g. La Meva Salut,
in Catalonia), which, is largely unstructured thus very
difficult to reuse. Therefore, citizens’ ownership of the data
becomes very difficult to implement in practice.
Nonetheless, transparent access to health
information presents challenges in the governance of the
same. New mechanisms need to be developed to protect
citizen confidentiality and privacy, while ensuring free
access to information for the benefit of the common good.
Advances in the field of genomics show there is scope for
developing new infrastructures, resources and policies
that promote the exchange of data for the common
good9
. The Human Genome Project10
, Encode Project11
,
Personal Genome Project12
, Wellcome Trust Case Control
Consortium13
, and Spanish BioBancos Network14
are some
examples. Most of these initiatives, however, consider
citizens as passive agents, since they are not involved in
the process beyond the informed consent they sign when
agreeing to donate their data15
.
These are some of the questions that have driven
the study described in this report:
9 Kaye, J., & Hawkins, N. (2014). Data sharingpolicydesign-
forconsortia: challengesforsustainability. Genome medicine, 6(1), 1.
10 https://www.genome.gov/
11 https://www.encodeproject.org/
12 http://personalgenomes.org/
13 https://www.wtccc.org.uk/
14 http://www.redbiobancos.es/
15 Woolley, J. P., McGowan, M. L., Teare, H. J., Coathup,
V., Fishman, J. R., Settersten, R. A., ... &Juengst, E. T. (2016).
Citizenscienceorscientificcitizenship? Disentanglingthe uses of pub-
licengagementrhetoric in nationalresearchinitiatives. BMC medical
ethics, 17(1), 1.
The data heals
Improve diagnosis
The Haematological Malignancy Research
Network (HMRN) is a collaboration between
epidemiologists, a centralized diagnostic
service and 14 hospitals that are capturing
detailed data on treatments, responses and
outcomes of clinical trials of every patient
with haematological cancer in Yorkshire and
Humberside. Since 2004, data of more than
20.000 patients has been registered. These
data have provided a very valuable insight
into potential pathways to diagnosis, gaps
and inaccuracies in physician guidelines about
symptoms, variations in treatment responses and
a socioeconomic survival effect.1
1 Haematological Malignancy Research Network

9
SALUS.COOP
CHAPTER 1
THIS STUDY
• Is it possible to involve citizens more
actively in the decision-making
process related to the use of their
health data?
• What kind of governance models
can be developed to give citizens
ownership and control of their health
data?
• How can we encourage data sharing
that benefits citizens, health
professionals, researchers, healthcare
providers and companies willing to
offer services/products?
“Today, the clinical and genetic information
of cancer patients is held in a variety of places:
academic medical centers, community hospitals,
disease-specific foundations, pharmaceutical
companies, and the government. There is very
little sharing of data among these institutions”
Hamermesh R. and Giusti K., One Obstacle to Curing
Cancer: Patient Data Isn’t Shared. Harward Business
Review, 28 Nov. 2016
“Today the public invests heavily in cancer
research through federal tax dollars, but the
current academic publishing environment hampers
innovation and discoveries. Research articles
are hidden behind paywalls, and delayed from
release by long embargoes. Research data remain
unavailable, or are restricted from being machine-
readable to allow deeper analysis. An alternative
system, where all publicly-funded cancer research
and data are required to be shared, would allow
researchers to unlock their content and data for re-
use with a global audience, and co-operate towards
new discoveries, analysis, and cancer treatments.”
Ryan Merkley CEO, Creative Commons, 2016 (Wiki
Creative Commons)
The data heals
Improve prevention
A team from the Houston Methodist Research
Institute has developed software that extracts
medical information from clinical reports, patient
scan records, and mammography results. It builds
risk estimation models to reduce unnecessary
biopsies.1
1 Patel, T. A., et al.,(2016). Correlating mam-
mographic and pathologic findings in clinical de-
cision support using natural language processing
and data mining methods. Cancer.

10
SALUS.COOP
CHAPTER 1
THIS STUDY
1.2 Vision and objectives
In accordance with the current context, it is believed
there is a need to promote new forms of health data
management that recognize the right and ability of
citizens to decide when, how, what, why and with whom
they want to share their health data. With the support of
the Mobile World Capital Barcelona Foundation, a study
has been developed, which aims:
To explore the potential
for developing a
citizen-driven model of
collaborative governance
and management of health
data, which enables
citizens to collectively
share data and therefore
accelerate research and
innovation in healthcare.
In order to address this objective, a model
(hereinafter, the Salus model) is envisioned (Fig. 1). It takes
three main groups of actors into account:
Citizens: the legal owners of their health data, which
are stored in several databases.
Data keepers: the owners of the databases where
citizens’ health data are stored. Some potential data
keepers include private and public health centers, smart
device companies (wearables, apps).
Data users: the parties interested in accessing the
data for different purposes. Possible recipients include
researchers, companies and public administrations.

11
SALUS.COOP
CHAPTER 1
THIS STUDY
Data keepers Data usersCitizens
Data users
for offering personalized services
• Service companies
• Health companies
• Startups
• Medical associations
• Administrations
• Others
Data users
for conducting research
• Research centers
• Universities
• Research units in companies
• Others
Cooperative
• Public health centers
• Private health centers
• Apps/ wearables/ devices
• Personal data
• Others
FIGURE 1. Salus model projection
This project is the first step towards understanding
the viability of implementing a model such as Salus. The
aim of this study is not to be exhaustive but rather to
inspire new ways of thinking and framing the promising
interplay between citizen’s health data and participation.
The focus of this research is on understanding
how the model can be applied in a specific case: breast
cancer. This disease has been chosen for several reasons:
(i) in Catalonia there is an active and empowered patient
association, (ii) there are several research centers that
are renowned worldwide and (iii) there is a high social
awareness about this disease.

12
SALUS.COOP
CHAPTER 1
THIS STUDY
1.3 Methods
The overall objective of this project has been
achieved by:
1. Gathering opinions and suggestions from experts
representing each identified group
2. Analyzing existing initiatives on health data
sharing
3. Designing a possible governance model that
recognizes citizen ownership of health data.
This study was developed by following a qualitative
approach to data collection16
. The following sections
describe the research methods followed.
Semi-structured interviews
Semi-structured interviews were conducted with
representatives of each of the three groups: citizens,
data keepers and data recipients. Interviews were aimed
at understanding people’s interests and perceptions of
the envisioned Salus scenario. Before the interviews, a
brief presentation describing the Salus model was sent
to interviewees. Each interview began by explaining the
Salus model, and then questions were asked about (i)
the perceived benefits, risks and barriers, (ii) the possible
conditions that could be established, and (iii) any other
16 Gill, P., Stewart, K., Treasure, E., & Chadwick, B. (2008).
Methods of data collection in qualitative research: interviews and
focus groups. British dental journal, 204(6), 291-295.
relevant issue perceived by the interviewee.
To determine the sample group of interviewees
a snowball sampling approach17
was followed.
Representatives with an involvement in breast cancer,
were identified. The selected sample was made up of 24
people, including citizens’, representatives of the data
keepers and data recipient, and other experts in the fields
of ethics and technology. For the citizen group, members
of breast cancer associations and other associations,
such as fibromyalgia and chronic fatigue patients, were
also selected. In both the data keeper and data recipient
groups, participants included professionals from several
hospitals and institutions involved in cancer treatment
and research, such as the Institut Català d’Oncologia
(ICO), Instituto Oncológico Baselga (IOB), Hospital Sant
Pau, Hospital Clínic and Hospital del Mar. From the group
of experts, bioethics researchers from the University of
Barcelona, bioinformatics researchers from Bioinformatics
Barcelona, and open data experts from Barcelona Open
Data Foundation were interviewed.
Interviews were conducted in person or by
telephone. Notes were taken during each interview and
transcribed for subsequent analysis. Data were analyzed
by two coders through deductive thematic analysis18
,
which consisted of reading the notes, creating reference
17 Goodman, L. A. (1961). Snowball sampling. The annals of
mathematical statistics, 148-170.
18 Braun, V. and Clarke, V. 2006. Using thematic analysis in
psychology. Qualitative research in psychology, 3(2): 77–101. doi:
10.1191/1478088706qp063oa

13
SALUS.COOP
CHAPTER 1
THIS STUDY
codes (tags), identifying themes by grouping related codes
together, and reviewing themes in an iterative model
by checking the entire data set. Themes were discussed
during group sessions with the research team in order to
validate and refine them (Figure 2).
The core themes that emerged from our analysis,
and that are used to present the results in section 2, are as
follows:
Universal benefits: innovation (research, business),
provision (prevention, service management).
Terms for data donation: control (over access, over
use), transparency and communication, collective benefits,
anonymity and security
Barriers: entry barriers for citizens (motivation, lack
of understanding, access), barriers between the parties
involved (doctor-patient, medical practices, distrust
among parties), technological barriers (unstructured data,
non-aggregated data, natural language).
FIGURE 2. Map with initial themes that emerged from the analysis of data interview. Used in the group
session with the research team.
Desk Research
In order to better understand the envisioned
scenario in relation to the current health data
ecosystem, information on existing initiatives led by
governments, companies, research centers, citizens
and others were collected. Information was gathered
in scientific magazines (e.g. Nature, Nature Genetics),
Journals (e.g. BMC Medical Ethics, Genetics in Medicine,
Genome Medicine, Nature Biotechnology), conferences
proceedings (e.g. Workshop on Sharing Clinical Research
Data 2013) and the Internet (e.g. ProPublica, websites of
the identified initiative).

14
SALUS.COOP
CHAPTER 1
THIS STUDY
Validation sessions
The analysis of interview data, as well as the analysis
of existing initiatives, was presented and confirmed in
two validation sessions. 34 experts participated in these
sessions, of which 12 were interviewed in advance. Each
session started with a presentation made by the research
team in which the results of the interview analysis
and a preliminary analysis of existing initiatives were
presented. In each session, participants were grouped
Beneﬁts
Risks
Cooperative
Maria’s data is stored in differ-
ent databases. Maria has access
credentials and shares them
with the cooperative under
speciﬁc conditions that she
established.
Private
Maria stores her health data on
a platform owned by a private
company (e.g. Apple health or
other PHR apps)
Public
Maria’s data is stored in public
databases. Maria has creden-
tials to access her data (e.g. La
Meva Salut)
Individual
Maria’s health data is
stored in her personal
hard disk
FIGURE 3. Matrix used in the validation sessions.
into three teams and asked to fill out a table (Fig. 3)
with the benefits and risks they perceived in relation to
different governance models (i.e. individual, public, private
and cooperative). At the end of the activity, each team
presented the resulting table to the other teams. The
session concluded with an open group discussion aimed
at summarizing the main conclusions. Subsequently,
the tables filled out during the validation sessions were
analyzed by the research team to highlight common topics
and concerns.

16
SALUS.COOP
CHAPTER 2
RESULTS
This chapter presents the results of the research conducted as part of this project.
Section 2.1 presents an analysis of existing data sharing initiatives that are related to
the goal of this project. Sections 2.2 and 2.3 detail the results of the field study, namely
the interviews and validation sessions. Section 2.4 summarizes the results by outlining
the four aspects that have been identified as fundamental to creating a citizen-driven
governance model for health data, and that have been used for designing the Salus
model presented in chapter 3.
2.1. Mapping existing health data
sharing initiatives
It is well-known that the healthcare sector needs
data on a large number of people in order to make
advances in medical research. This need has generated
a demand for involving citizens in order to encourage
them to make their data available1
. There are numerous
initiatives that have promoted data sharing for research,
and they differ in terms of: mission statements,
approaches to citizen involvement, data policies, funding
schemes, governance model, and promoters, among other
aspects. Within these initiatives, the focus has been on
the elements that are considered most relevant to the
project’s objective: the citizen’s role and governance
1 Woolley, J. P., McGowan, M. L., Teare, H. J., Coathup, V.,
Fishman, J. R., Settersten, R. A., ... & Juengst, E. T. (2016). Citizen science
or scientific citizenship? Disentangling the uses of public engage-
ment rhetoric in national research initiatives. BMC medical ethics,
17(1), 1.
models.
Five different roles for citizens, which range
from a more passive to a more active role, have been
identified. By governance model, it is meant the control
and management scheme related to the type of party
running the initiative. For example, an initiative led by a
government institution is likely to derive a governance
model that resembles the provision of public services. On
the other hand, an autonomous association of people will
require a non-hierarchical and participatory arrangement,
here referred to as collective direct democracy.
Table 1 shows a breakdown of the citizen’s role.
Table 2 details the type of participants who promote and
govern the initiative, which is a key element to defining
the governance model. Table 3 presents the correlation
between the two.

17
SALUS.COOP
CHAPTER 2
RESULTS
TABLE 1. Breakdown of the role of citizens in health research initiatives based on data sharing.
TABLE 2. Breakdown of governance models in health research initiatives
Role of citizens Characteristics Examples
Informed and
consenting
patients
People are asked to give
their consent to donate
their data for public
research.
Visc+ (Catalunya)
Cara.data (UK)
Genome Project (Estonia)
Contributors People proactively decide
to donate their health
data by, for instance,
uploading personal data
into platforms.
DataDonors
Open Humans
Consumers People share data with
companies who provide
them with health devices,
applications or services.
PHR apps (e.g. Microsoft
HealthVault)
Wellness Apps and other
devices (Fitbit, Garmin, etc.)
Personal service (e.g.
23andMe)
Partners People are informed about
the use of their data and
can participate in decision
making processes.
HealthBank
Midata.coop
OHDC
Prosumers People are provided with
tools that enable them to
be involved in research.
For instance, they can
provide information about
their disease through
social networking tools
(PatientsLikeMe).
Social networking services:
e.g. Patientslikeme
Promoters
Governance
agreement Examples
Governmental
institutions
Public Visc+ (Catalunya)
Cara.data (UK)
Genome Project
(Estonia)
For profit
companies /
Corporate
Private PHR apps (Microsoft
Health Vault)
Wellness Apps and
other devices (Fitbit,
Garmin, etc.)
23andMe
Patientslikeme
Not-for-profit
organization: asso-
ciations, founda-
tions, NGOs
Collective -
representative
democracy
DataDonors
Open Humans
Autonomous
public associations
(e.g. cooperatives)
Collective
- direct
democracy
Health Bank
Midata.coop
OHDC

18
SALUS.COOP
CHAPTER 2
RESULTS
Although all the initiatives analyzed share the same
objective (engaging citizens in biomedical research)
the way that citizen’s involvement is conceptualized
and articulated varies hugely. Our analysis shows that
privately held initiatives mainly focus on providing
citizens with products or services, which results in new
data being generated (e.g. genomic data (23andMe),
social networking conversations (PatientsLikeMe), and
sensor-generated data (FitBit)). These data are used in
business-driven research projects that result in patents,
products or new drugs from which (only) the company
derives a financial profit. This is the case, for instance,
of 23andMe, a company that provides customers with
genetic information, which has then been used to
obtain patents. This stirred up controversies about the
extent to which “any (private or public) organization
involved in research that relies on human participation,
whether by providing information, physical material,
or both, needs to be transparent, not only in terms
of research goals but also in terms of the strategies
and policies regarding commercialization”2
(p.382).
Other initiatives rely on a representative governance
model (e.g. associations, foundations), wherein citizen
participation is conceptualized in terms of proactive data
donation to research projects decided on by boards of
directors. Initiatives driven by public institutions tend to
2 Sterckx, S., Cockbain, J., Howard, H., Huys, I., & Borry, P.
(2012). Trust is not something you can reclaim easily: patenting in
the field of direct-to-consumer genetic testing. Genetics in Medicine,
15(5), 382-387.
conceptualize participation in terms of passive patients/
citizens who have the civic duty to participate in public
TABLE 3. Relationship between governance models and the role proposed to citizens
Collective/
direct
democracyPublic Private
Collective/
repre-
sentative
democracy
Informed and
consenting
patients
Visc+
(Catalunya)
Cara.data (UK)
Genome Project
(Estonia)
Contributors Datadonors
Open Humans
Prosumers Patientslikeme
Partners HealthBank
Midata.coop
OHDC
Consumers PHR apps
(Microsoft
HealthVault)
Wellness Apps
and other
devices (Fitbit,
Garmin..)
23andMe
Open Humans
Personal
Genome Project

19
SALUS.COOP
CHAPTER 2
RESULTS
research3
. Participation is conceived in terms of data
donation, and no opportunities are given to citizens to set
research priorities and agenda.
In summary, both the representative and public
models leverage a rhetoric of altruism, while the private
model uses the appeal of the benefits offered to individual
users, for instance in terms of the service or product.
Contrary to the aforementioned models, a limited
number of newly established initiatives are trying to adopt
a more participative and citizen-driven governance model,
which consider citizens as actual partners and owners of
the initiative. This means that citizens can exercise the
right to an economic reward for profits generated by data
use, such as in HealthBank, and the right to participate in
decision-making processes, such as in MiData.coop. This
model of citizen direct governance is still in its infancy and,
to the best of our knowledge, only a few such initiatives
exist worldwide, and they have not yet been deployed in
full.
There is a need then to understand further how a
model of this type could be enacted, and what challenges
it might face. The study presented in this report aims to
contribute towards filling this gap.
3 Woolley, J. P., McGowan, M. L., Teare, H. J., Coathup, V.,
Fishman, J. R., Settersten, R. A., ... & Juengst, E. T. (2016). Citizen science
or scientific citizenship? Disentangling the uses of public engage-
ment rhetoric in national research initiatives. BMC medical ethics,
17(1), 1
2.2. Results of the interviews
2.2.1. Universal benefits
All the groups interviewed perceive clear benefits
in developing a system that facilitates access to health
data and promotes data sharing. More specifically, they
emphasized that access to data has great potential for
supporting healthcare service provision and boosting
innovation in this field.
PROVISION
In relation to the provision of services, data
providers highlighted the potential shared health data
holds for the provision of services in terms of disease
prevention and the development of personalized
treatments. The interviewees pointed out that these two
aspects, both provision and personalization, could imply
improved resource management and a reduction in costs
for the public health system, as well as an improvement in
the service provided to patients.
INNOVATION
In terms of innovation, all the interviewees
highlighted the potential of shared health data, especially
in terms of accelerating research into medicine. The parties
interviewed pointed out people’s altruistic instinct when

20
SALUS.COOP
CHAPTER 2
RESULTS
it comes to health problems, and their willingness to
cooperate either to cure their disease or to help future
generations.
“Patients with breast cancer have an altruistic
instinct that will lead them to provide data if they
feel that this action may help advance research into
the disease.” Patients Association
Finally, it should be noted that another topic that
emerged during the interviews was the potential for
new businesses to be developed as the result of shared
data, which could contribute to improving the current
healthcare service model.
“Data can create industry benefits in terms of
creating new products and services for citizens.”
Professional care and research.
2.2.2. Terms for data donation
As described above, the willingness to share health
data has generally been very positive among respondents.
Nonetheless, our analysis also highlights that interviewees
would not be willing to allow access to their data just for
the sake of sharing. Rather our analysis outlines the desire
to donate data under a series of conditions described
next:
CONTROL
Regardless of the general willingness to donate data,
there is a clear desire to control which data is shared, who
has access to the data, what will the data be used for
and who will benefit from such use. Some interviewees
pointed out that prior to any use of the data, permission
should be sought from the owner.
TRANSPARENCY AND COMMUNICATION
One of the conditions required to allow control
over the data is guaranteed transparency for the whole
process, from data requests to delivering the results.
Transparency goes hand in hand with clean and clear
communication, which is the basis for allowing conscious
and informed decision-making by data owners. Patient
associations emphasized the importance of being able
to know what is happening to their data at all times,
since one of their fundamental goals is to safeguard
the wellness of their members. A lack of proper
communication could put their trust at risk, and with it,
the participation of members.
“It is important to always provide clear
information and full transparency about how data
are being used, both at the start and end of the
process.” Patients Association
COLLECTIVE BENEFITS
Respondents believe that an economic
compensation to individuals who decide to give their data

21
SALUS.COOP
CHAPTER 2
RESULTS
may have a negative effect. They pointed out blood and
organ donation models as good examples, since in these
cases altruism prevails over any individual benefit to the
donor. Some interviewees outlined other non-economic
forms of compensation, such as individual compensation
in the form of services, or collective compensation in the
form of charity donations.
“There should not be an economic
compensation for releasing data. But whoever
receives the data could make an economic
contribution to a social or a research project. The
person who has provided the data should be able to
choose the type of project.” Patients Association
“It would be interesting to explore the
possibility of providing companies with data that
could be used to offer personalized services to
patients.” Health professional
There is a clear perceived risk among most
respondents regarding data being used for profit, which
would not benefit citizens and could even be used against
them. The most noteworthy case reported in several
interviews relates to health insurance, as the improper
use of health data by insurance companies may result in
insurance contracts that do not benefit the individuals.
Interviewees also highlighted that the sale of data for
profit may have a negative impact.
In summary, all interviewees agree that the use of
health data must fulfill the condition of generating a clear
and unequivocal collective dividend, from which society
as a whole can benefit. Scientific research is considered a
key feature to achieve this goal. Moreover, interviewees
highlighted three important aspects regarding promoting
research for the common good:
• The importance of encouraging open access
research publications.
• The importance of publishing all clinical trial
results.
• Free and responsible research that addresses the
real needs of society.
ANONYMITY AND SECURITY
According to all interviewees, in order to respect
citizens’ privacy it is very important to design a system
that can ensure data anonymity. Therefore, they pointed
out that the system designed must avoid other parties
being able to re-identify citizens through data crossing.
As some interviewees pointed out, when working
with health data it is not possible to state that data are
totally anonymous, since some data are unique (e.g.
genetic data) and as such it could be easily re-identifiable
through data crossing. Several respondents suggested
that different procedures should be applied depending
on the likelihood of re-identification. Moreover, accurate
information about the potential risk of re-identification
should always be provided. The probability of

22
SALUS.COOP
CHAPTER 2
RESULTS
re-identification depends, among other factors, on: (i) the
amount of data available with regard to each individual;
(ii) the number of individuals in the database; and (iii) the
number of people with the same pathology (rare diseases).
Another factor to consider is the security of data
storage systems. It should be robust enough to withstand
computer attacks, which are increasingly frequent in the
healthcare sector. As some interviewees pointed out, a
distributed system could ensure a higher level of security,
since data would not be centralized in a single database.
2.2.3 Barriers
ENTRY BARRIERS FOR CITIZENS
The parties interviewed identified possible barriers
that could have a negative impact on citizens when
making their initial decision to join the cooperative, and
which might mean the organization could end up having
access to data which is insufficient and biased (e.g. if only
very active and motivated citizens join the cooperative).
The amount of data and quality of the sample are key
factors to ensure anonymity and have a valuable sample
that can be used for research.
Motivation
The fact that all of the interviewees were highly
motivated to participate in the proposed model must not
be taken as reflection of reality. Interviewees highlighted
the importance of motivating those citizens that are not
implicitly motivated to participate. In this regard, the
participants of the validation sessions remarked on the
need to develop individual incentives and rewards for
our system. For example, they proposed the possibility of
offering services to citizens that could help them manage
their health, such as Personal Health Records (PHR).
This might be a strategic aspect, taking into
account that there are a growing number of PHR health
apps. Citizens might be interested in uploading their
data to these apps in exchange for services that can
help them manage their health. However, none of the
already existing PHR allow citizens to share their data in a
collective way; furthermore, many do not clearly explain
what happens to the uploaded data, which is not aligned
with the aforementioned condition of a collective benefit.
Help to understand
Nowadays there is a widespread lack of knowledge
about how the research system works, and what the
data loops are. This means that the benefits of donating
health data may not be evident to citizens, which in turn
may have a negative impact on the participation of some
citizens, who may not understand the value of the model
proposed.

23
SALUS.COOP
CHAPTER 2
RESULTS
“There is a risk of non-participation if there
is not a clear evidence of the benefits related to
donating. Information is essential in this case
and can help mitigate any possible doubts and
concerns.” Health professional
Access
Interviewees believe there might be some difficulties
in accessing certain groups. For example, a lack of
technological knowledge among certain citizens may
have an impact on their levels of participation, as some
of the informational and decision processes will only
take place in digital contexts. Also, it might be difficult to
access people who are not very proactive and motivated.
This might be the case, for example, for citizens who are
affected by a specific illness but are not members of the
existing patients association.
BARRIERS BETWEEN DIFFERENT PARTIES
Some other barriers are related to what might occur
between the different parties involved in the system.
These barriers might arise given that: i) the proposed
model changes the current relationship between some
parties (e.g. doctors - patients), ii) some medical practices
might be affected, or iii) there might be some mistrust
towards the implicit objectives of some of the parties.
Perceptions of the possible changes in the doctor –
patient relationship
Health professionals consider that having access to a
large amount of data related to their health could cause a
certain level of anxiety among citizens. This poses the risk
that some of the information could be misinterpreted, or
that citizens might carry out erroneous self-diagnoses.
“The result of an analysis, prior to a diagnosis,
could be misinterpreted and generate a feeling
of anxiety among patients. While understanding
that patients have the right to this information, we
need to be aware that this vulnerability should be
managed.” Health professional
This could affect the relationship between the
health professional and the patient, given that there is
currently asymmetric information between these two
parties (e.g. the doctor has access to more information
than the patient). If the proposed system is put in place,
the asymmetry of the information will be reduced more
quickly than at present (websites, mobile, devices, etc.).
On the other hand, interviewees highlight
that health professionals might feel judged on the
information provided to citizens. Patients could ask them
for explanations of certain diagnoses and treatments
recommended by other health professionals.
Perception of possible changes to medical practices
Interviewees highlighted that the new system for
data sharing should not increase the workload of health
professionals. They are concerned that the new system

24
SALUS.COOP
CHAPTER 2
RESULTS
might have a negative impact on their work (e.g. forcing
them to introduce more data in the system or making
them complete more application forms, forcing them to
reduce the quality and quantity of the time they spend
with patients). Thus, this model must make sure to include
user-centred-designed technologies that do not have a
negative impact on in-person healthcare services.
Distrust between actors
Interviews have shown that there is certain mistrust
towards some of the parties involved in the system. In this
sense, our results confirm what other studies have already
shown4,5,6
: citizens are less willing to share data with
private companies because they fear that pharmaceutical
companies or other private institutions might make non-
ethical use of their data. On the other hand, interviewees
also recognize that these parties play a critical role, as they
provide medicine and offer services that improve people’s
health. The main challenge in this regard is to ensure that
the results obtained thanks to citizens’ data are used
for the common good. The price of medicine, research
agendas, and the transparency of results, are perceived as
the key elements to ensure this objective. These results
reaffirm the importance of putting access conditions
4 KPMG Survey, 2015. UK adults trust GPs with digital health-
care data but don’t want data from their fridges to be shared
5 Weitzman, Elissa R., et al. “Willingness to share personal
health record data for care improvement and public health: a survey
of experienced personal health record users.” BMC medical informat-
ics and decision making 12.1 (2012): 1.
6 Personal Data for the Public Good – Health Data
Exploration Project. 2014. Robert Wood Johnson Foundation
in place for the data to create a collective benefit, as
described in the previous section.
TECHNOLOGICAL BARRIERS TO DATA
Some of the interviewees pointed out potential
technological barriers that could hinder the development
of the system. The biggest barrier is linked to the fact that
most health data are currently saved on a non-structured
format. Blood analytics, for example, are delivered in a PDF
format. Furthermore, some non-structured data contain
natural language (e.g. reports written by the doctor). These
documents need to be interpreted clearly before they can
be classified.
It is worth pointing out that some interviewees
claimed that these barriers are likely to be overcome in the
short term, thanks to rapid technological developments in
the field.
2.3. Results of validation sessions
The analysis of existing initiatives (section 2.1) has
revealed a wide variety of governance models in data
management systems. Two validation sessions with 34
experts have been conducted in order to explore the
perceived benefits and risks of four types of governance
models: individual, private, public, and cooperative. A
scenario was presented to help participants envision
each of the models. The results can be summarized in the
following diagrams:

25
SALUS.COOP
CHAPTER 2
RESULTS
PRIVATE
Maria stores her health data on a platform owned by
a private company (e.g. Apple health or another PHR
app)
• Direct incentive to individuals: people are
motivated to upload their health data on com-
pany platforms because they receive services
and products in exchange that enable them to
access and manage their data in an easy way.
• Technological capacity: companies are per-
ceived to have high level of technological capa-
bility. People feel they can rely on the technolo-
gy used.
• Ultimate purpose of data: people have little or
no knowledge about what companies do with
their data. The terms of use that people sign up
to use the product are often difficult to under-
stand. One key point is that some users might
not be aware about secondary use of their data.
• No interoperability: generally speaking, compa-
nies do not have access to the medical records
stored in the databases of different health
centers.
• Risk of losing data: despite the general trust in
the technological capability of companies, there
is still the risk of potential data loss.
Scenario:
Benefits:
Risks:
INDIVIDUAL
Maria’s health data are stored on her personal hard
disk
• Autonomy: individuals do not depend on
others to decide what to do with their data.
• Control: individuals have maximum control
over their data, thus there is minimal fear about
potential misuse of the same.
• Limited contribution to research: individual
data are not very valuable for research, since the
value of personal data lies in its aggregation.
• Data loss: if data are lost, individuals cannot
complain to third parties because data were
held on personal storage.
• Isolation: since there is not a trusted party
acting as informant, individuals willing to donate
data for research do not know how to do it.
Moreover, individuals may receive requests from
third-parties about accessing/using data, and
might not know how to react as they do not
have adequate information.
Scenario:
Benefits:
Risks:

26
SALUS.COOP
CHAPTER 2
RESULTS
COOPERATIVE
Maria’s data are stored on different databases. Maria has a login to access
her data, and shares them with the cooperative under specific conditions
that she has established.
• Integrative model: the cooperative can integrate data from different
sources, both public and private data keepers (hospital, private clinic,
and wearables). This would allow different type of information to be
integrated and correlated, including habits, sensors, and medical records.
• Guided decision: people willing to donate their health data can be
informed and assisted by the cooperative, which provides them with the
information needed to take an informed decision about their data. People
who trust the cooperative can delegate decisions to it.
• Independent from political changes: unlike public institutions, the
cooperative does not depend on political changes.
• Lack of individual incentives: there might be a lack of interest if people
cannot clearly see how they personally benefit from participating in this
model.
• Lack of critical mass: motivating a large number of people to participate
in the cooperative might be challenging. It might thus be difficult to reach
a critical mass of participants, which is key to having a representative
sample for research studies.
• Complexity of governance model: the governance model is viewed as
being complex, especially with respect to building trust, transparency,
and ensuring public interests.
• Difficulty in accessing the data: the infrastructure of data systems is
complex, which means it might be challenging to access data from differ-
ent sources. The data transfer process by public authorities might also be
slow due to bureaucracy and legislations.
Scenario:
Benefits:
Risks:
PUBLIC
Maria’s data are stored on several databases owned by
public institutions. Maria has a login to access her data (e.g.
La Meva Salut).
• Large volume of data: public institutions potentially
have access to many medical records, and to all of the
data stored on public health centers. This large volume
of data would make it possible to conduct popula-
tion-based research.
• Guarantor of data use: public institutions are per-
ceived to be good guarantors of data use, as their
objectives are in the public interest.
• Political changes: public institutions are subject to po-
litical changes, which might slow down, or even stop,
initiatives promoted by previous governments.
• Little capacity of investment: little capacity for
investment might lead to technological solutions being
developed that are not sufficiently innovative, or there
might be a lack of incentive for accelerating research in
healthcare.
• Risk of losing data: public institutions are often
accused of being rigid, slow to adapt, and crippled by
bureaucracy. The perceived risk is that this rigidity and
slowness might affect the overall management of the
data sharing system, and could eventually result in a
missed opportunity to make the most of the data to
accelerate research.
• No direct decision from the citizen: citizens do not
take part in the decisions made regarding the second-
ary use of health data.
Scenario:
Benefits:
Risks:

27
SALUS.COOP
CHAPTER 2
RESULTS
In conclusion, the validation sessions highlighted
the importance to carefully consider the following two
aspects:
• Individual incentives: besides ensuring
public interests, it is important to give concrete
benefits to individuals in order to motivate
them to participate. Individual incentives are
the driving force that makes possible to reach a
critical mass of participants and make worthwhile
contributions to science.
• Flexible and transparent collective
governance: groups of people have greater
control over their data than individuals acting
alone. The collective governance model
should ensure transparency and participation
in decision-making processes. This should be
enabled by a flexible participation infrastructure
that allows for informed decisions and a
delegative democracy. Transparency fosters
trust, which is key to encouraging people to
participate in the model.
2.4. Conclusion: The pillars for
citizen-driven governance of the
health data
In summary, the results of our study have
highlighted four fundamental principles that should be
considered the pillars of any citizen-driven governance
model for health data management.
Conditional donation
Citizens have the right to decide under which
conditions they want to donate their health data.
The transfer of data would be made in the form
of a donation. This donation would be motivated by
the altruistic spirit of citizens and would prioritize the
common good over any individual economic benefit. In
any case, certain conditions should be fulfilled by the data
receivers in order to respect the altruistic spirit of donors.
Citizens want to know who will be using their data and
what for. With this information, they will be able to decide
if they want to donate their health data or not. In this
regard, clear and transparent communication throughout
the process is a fundamental aspect of donation.

28
SALUS.COOP
CHAPTER 2
RESULTS
Collective benefits
The use of data by any parties involved should
provide a clear benefit to society.
The system needs to ensure that the use of data
by any of the parties has a clear and unequivocal benefit
for society as a whole. For instance, the outputs of the
research conducted by using citizens’ data could be made
universally available under an open licence. Moreover,
citizens should be given the possibility to identify research
priorities and set research agendas. The specificities of any
governance protocol that allows a given system to ensure,
safeguard and produce a collective benefit must be further
investigated.
Motivational incentives
Incentives should be given to individuals to
motivate them to donate their data. Incentives
should not be monetary.
We need to find an incentives system that can
motivate as many citizens as possible. This would enable
access to a data sample that is diverse and representative
enough to be valuable for researchers. Nevertheless, the
incentives should not be monetary for the individual, so
that they do not hinder the common good. Incentives
could be provided, for example, in the form of services to
members of the cooperative.

29
SALUS.COOP
CHAPTER 2
RESULTS
Management of rights
A participatory governance model should be
deployed to guarantee the collective benefits of
data use and citizen control over the fate of
their data.
A collective governance model that enables the
access to and management of decentralized databases
is required, while applying the conditions established by
data owners. The governance model would manage: i) the
conditions established by members of the cooperative,
ii) the credentials required to access the data and iii)
requests from data receivers, which must comply with the
principles established by the cooperative.

CHAPTER 3
SALUS: towards citizen governance
and management of health data

31
SALUS.COOP
CHAPTER 3
SALUS: TOWARDS CITIZEN GOVERNANCE AND MANAGEMENT OF HEALTH DATA
Following the results of our study, a model for citizen-management and governance of
health data is proposed. The model responds to the following needs:
• To empower individual citizens so that they
can control what happens to their data (Pillar 1:
conditional donation) and to incentivize them
to donate their data (Pillar 3: motivational
incentives).
• To aggregate health data from large groups of
citizens, who govern the aggregated dataset as
collective and according to the terms agreed by
the group itself (Pillar 4: rights management).
• To foster research and innovation in healthcare
through the use of the aggregated dataset, by
allowing citizens to identify research priorities
and set research agenda (Pillar 2: collective
benefits).
In this section a model that aims to meet these goals
is presented. It includes the following frameworks:
• Governance and relational framework
• Technological framework
• Legal framework
It is worth emphasizing that this study represents
an initial exploration that outlines possible scenarios that
could inspire new ways of thinking about and framing the
interplay between citizen’s health data and participation.
For this reason, this report does not present a complete
and exhaustive description of the proposed model. It
outlines some of the aspects that should be taken into
account in successive phases of this study, and that
might contribute towards achieving this project’s overall
objectives.
3.1 Governance and relational
framework
The proposed model is made up of a series of
different actors (section 3.1.1) and the relationships
between them (sections 3.1.2 to 3.1.5).
The relationships between them are enacted
through i) the set of values exchanged between them
and ii) a set of agreements that define these exchanges.
In terms of the exchange of values, three types of value
flows have been identified: i) data flows, ii) service flows
and iii) economic flows. The data flow represents the main
pillar of the system. The service flow acts as an incentive
mechanism that motivates the different parties in the
model to participate and get engaged. The economic
flow allows the model to be economically sustainable.
Overall, these three flows make up an ecosystem wherein
assembled datasets are the main assets of a system based
on data economics.
Five different types of agreements have been
identified to regulate the relationships among parties:
i) ethical agreements, ii) membership agreements, iii)
data transfer agreements, iv) service agreements and v)

32
SALUS.COOP
CHAPTER 3
exploitation agreements. Each one is aimed at the different
parties and flows in the model. Overall, these agreements
make sure that activities arising from the aforementioned
flows comply with the fundamental principles established
by the cooperative.
It is important to note that all the decisions
regarding the specific content of these flows and
agreements will be made collectively, through the internal
governance processes of the cooperative. The parties,
Cooperative
Public health centers Universities
Research centers
Clinical research companies
Others
Private health centers
Others
Data keepers Data users
Providers of services
to data keepers
(aggregation and
structuring of the data)
to data users
(analysis, visualization,
modelization...
Providers of services to
cooperative members
(PHR, wearables...)
ethical agreement
Data keepers
to data keepers
cooperative members
flows and agreements shown in the next section are
intended as a preliminary proposal.
3.1.1. Ecosystem of actors
Our proposal requires an ecosystem formed by
groups of different actors in order to be viable. Each party
provides a different value to the model, and this value
could either be in terms of data, infrastructure, services, or
research.
FIGURE 4. Ecosystem of actors

33
SALUS.COOP
CHAPTER 3
THE COOPERATIVE
The results of our study pointed out the need to
develop a collective structure that would ensure: (i)
transparency in data processes and data usage; (ii) the
participation of citizens in decision-making processes; (iii)
shared ownership of any value generated. A cooperative
model could and should allow us to meet these goals.
Value provided: The cooperative is the mediating
agent between the different parties. The cooperative
proposes and implements governance and management
systems, which are used to articulate the relationships
between parties. By building upon principles of liquid
democracy enabled by new technologies (e.g. smart
contracts1
), the cooperative will have a governance
structure that will ensure a dynamic, transparent and agile
decision-making process.
COOPERATIVE MEMBERS
This group includes citizens who join the
cooperative.
Value provided: They give the cooperative the legal
authority to access data on their behalf, under the terms
set by the individual members, and only when fulfilling the
objectives of the cooperative.
Value received:
• They can vote in the cooperative assembly,
1 O’Day, C. Liquid Democracy and Emerging Governance
Models. Consensys Media. 24 August 2016
which enables them to control the fate of their
data.
• They can actively contribute to set research
agendas by (i) donating their data to specific
research projects and (ii) taking part in decision-
making processes through which the cooperative
chooses which research projects should the data
be granted to.
• They can access services and products (PHR,
apps, etc.) offered by service providers
accredited by the cooperative.
DATA USERS
This group includes all the parties who are interested
in accessing the cooperative dataset for research and
innovation purposes.
Value provided:
• They conduct research by making use of the
cooperative dataset, thus providing valuable
information to the cooperative itself and the
society as a whole.
• They develop products and services by making
use of the cooperative dataset, thus accelerating
innovation in healthcare.
Value received: Access to an assembled set of
health data from various sources under the economic
conditions set by the cooperative.

34
SALUS.COOP
CHAPTER 3
DATA KEEPERS
This group includes all the parties that own the
databases storing citizen health data.
Value provided: Access to the health data of
cooperative members.
Value received: The data are returned to data
keepers in a structured and aggregated format. In this way
data keepers can progressively improve the quality of their
databases.
SERVICE PROVIDERS
This group of parties are instrumental to put the
service flow in place (described in section 3.1.3.), providing
values -services- to all the parties involved in the model.
In order to be a service provider, companies must be
accredited by the cooperative. The accreditation is aimed
at ensuring the quality and transparency of the services
provided, as well as the work processes. Three main
groups of service providers were identified, depending on
the parties they provide the service to.
Value provided:
• To cooperative members: They provide health
related services or products to cooperative
members, giving them a material incentive to join
the cooperative.
• To data keepers: They provide data from
cooperative members to data keepers in an
aggregated and structured form. This helps
obtaining an important set of correlated
information that ensures patients and other
parties in the health system are offered more
effective services and treatments.
• To data users: Upon request by data users,
they can provide users with data processing
services, such as visualization, modeling, features
extraction, analysis, and more.
Value received: The cooperative offers service
providers with a strategic option to enter a new market
and gain access to a range of potential customers, e.g.
patients, health centers, research centers.
Being a Salus accredited company could have a
positive impact on a company’s image.

35
SALUS.COOP
CHAPTER 3
FIGURE 5. Data flow.
Cooperative
Research centers
Clinical research
companies
Others
Others
ethical agreement
1. Request
2. Collective decision
3. Request
4. Extraction
Providers of
services to
data keepers
cooperative members
to data users
5. Preparation
6a. Delivery6b. Delivery
2. Collective decisions
Internal bodies in the
cooperative review the
grant application and decide
whether it complies with the
cooperative’s statute. If it
does, a price is automatically
established for data access,
according to the criteria of
the cooperative assembly.
Internal governance processes
allow individual members to
decide whether to donate
data to each particular grant
application.
1. Request
A group of researchers submit
an application to request
a certain set of data. The
application contains detailed
information about the research
study: data usage, access and
storage, funding, expected
outcomes, licensing, etc. If
approved, researchers sign
a contract wherein they
declare their liability in case of
negligent data usage.
6b. Delivery
Aggregated and structured
data are returned to the data
keeper - health center.
5. Preparation
Data undergo a process of
anonymization, aggregation
and structuring, carried
out by the technological
providers associated with the
cooperative.
6a. Delivery
Raw anonymized data are
delivered to researchers.
Additional services, such as
data visualization, modeling
and analysis, can be provided
by accredited technological
partners.
3. Request
4. Extraction
The cooperative requests
the data from health centers,
in accordance with the
agreements entered into place
between the two parties.
Technological providers
accredited by the cooperative
extract data from health center
databases, in accordance with
the agreements between
the parties. Initially the data
is neither structured nor
aggregated.
3.1.2. Data flow
The data flow links
up all the parties in the
model and serves as the
basis for their relationships.
The process envisioned is
laid out in six steps. Some
of these steps could be
automated or accelerated,
while remaining transparent in
the process. Further research
is needed to determine how
to balance transparency and
automation through the right
technological solutions and
governance processes.

36
SALUS.COOP
CHAPTER 3
Cooperative
Research centers
Clinical research companies
Others
Others
ethical agreement
Providers of
services to
data keepers
cooperative members
to data users
Analysis,
visualization,
modelization
Aggregation and
structuring of the data Data use and access services
(PHR, wearables...)
FIGURE 6. Service flow.
3.1.3. Service flow
One of the key
findings of our study is
that incentive mechanisms
should be established in
order to motivate parties
to participate in the model.
The service flow in the Salus
model is aimed at reaching
this goal. Accredited
companies are in charge of
developing the services to be
provided to different parties.
Our findings highlight that citizens
might be willing to participate if they
perceive personal benefits. Services
provided to cooperative members could
be aimed at helping them access and
manage their integrated health records
by providing them with the right tools
(e.g. Personal Health Record apps). Other
potential services could facilitate self-
management of diseases. Purchasing
these services as a group would imply a
reduction in price.
Our findings pointed out that accessing data from different
data keeper systems might be challenging, due to both
technological and bureaucracy barriers. It is also well-known
that data keepers, such as public health centers, often face
challenges in keeping data structured and aggregated.
The proposed model aims to address these challenges
by rewarding data keepers with structured data from the
cooperative members. In this way, as the Salus model
progressively takes shape, data stored in data keepers’
databases will be gradually structured. Data keepers should be
motivated to proactively participate in the model and speed
up their administrative processes.
Data (pre-) processing services
could be offered to data users
on request. Some examples of
these services include cleaning,
features extraction, modeling and
visualization. These services are
aimed at increasing the quality of
the data offered.
Services for data keepers: Services for data users:Services for members:

37
SALUS.COOP
CHAPTER 3
3.1.4. Economic flow
The system of relationships
also requires an economic flow
that will maintain the structure
of the cooperative, and payment
for the services provided by the
associated service providers.
The economic flows
identified in this model are
structured around two key
elements (i) an internal currency,
named SalusCoin, and (ii) a
variable price for access to
data according to a set of key
variables.
Cooperative
Public health centers
Subsidized projects
by the cooperative
Universities
Research centers
Others
Others
ethical agreement
Data keepers
Data users
Providers of
services to data
keepers
cooperative members
to data users
Payments to
service providers.
Payments to
service providers.
Payments to
service providers.
With the fund generated
by the SalusCoins, the
cooperative could decide
to fund studies.
Depending on the key indica-
tors deﬁned by the coopera-
tive, data users might be
asked to pay a variable price
to access data.
FIGURE 7. Economic flow
INTERNAL CURRENCY: SalusCoin
SalusCoin is the internal currency developed by
the cooperative, which gives value to data and the
participation of cooperative members. Our research
showed that the organization should be a non-profit
entity and that members of the cooperative should not
receive personal monetary incentives. For this reason, the
creation of an internal currency was proposed to allow the
cooperative to reward the contribution of all the parties
involved in making the model possible.
VARIABLE PRICE
In order to achieve our goal of having an impact on
medical research agenda, the cooperative could promote
certain data uses over others. To this end, the cooperative
might decide to offer different economic conditions
to different data users, for instance, to academic or
commercial users. The variable price that the cooperative
applies to data users is aimed at having impact on the
research agenda. For instance, better conditions can be
provided to research projects that are more aligned with
the values of the cooperative, and charge a higher price
for projects that are considered less relevant by members

38
SALUS.COOP
CHAPTER 3
of the cooperative (e.g. research on rare diseases vs.
research on wrinkles).
In order for the cooperative to determine the price
to be applied, data users would be asked to specify
relevant information in their grant application. In general
terms, the price would be positive, negative or zero.
Positive price: When applying a positive price, data
receivers will have to contribute economically to the
cooperative in order to gain access to the data. This
amount would respond to the costs incurred by the
service providers to make the data accessible, or would be
higher if the data requested is not aligned with the values
of the cooperative.
Zero: If the price applied is zero, the cooperative would
share the data at zero cost to the data receiver.
Negative price: There might be certain cases in which the
cooperative wants to contribute to research in a specific
field, not only with data, but also financially. In these cases
the cooperative would invest SalusCoin in supporting a
specific research project (e.g. breast cancer).
REVENUE STREAMS
Taking into account the two key elements that
affect the monetary exchanges of the cooperative, diverse
revenue streams are envisioned.
Factors that affect the pricing decision
The cooperative defines the price to apply to data,
according to a set of variables. Possible variables are:
• Type of party requesting the data (e.g.
academic user, commercial user)
• Aim of the research project for which the data
will be used
• Partners involved in the research project
• The research project’s funding scheme
• Expected output
• Applicable licenses and publications policies
• Previous transactions with the cooperative
• Additional services (e.g. visualization,
modeling, etc.)
Payments made by data users: Depending on the price
range determined by the cooperative, data users might be
asked to contribute financially to the cooperative.
Fees paid by service providers for receiving
accreditation: In order to ensure the reliability of the
services provided and the security of the data processes,
the cooperative would accredit companies that want to
act as service providers. Being accredited by Salus would
provide a market opportunity for service providers, so
Salus will charge a fee to obtain the official accreditation.
Membership fee: Citizens who want to join the
cooperative will be asked to pay a membership fee.

39
SALUS.COOP
CHAPTER 3
A portion of this contribution will be placed in the
cooperative fund, and the rest will be provided to the
individual in the form of SalusCoin in their personal wallet.
INCOME DISTRIBUTION
The income received from the revenue streams
described above will be distributed in the following ways:
it will be used to cover the infrastructure and running
costs of the cooperative, and to create a fund to influence
the research agenda.
Infrastructure and running costs of the cooperative:
To cover the cost of the cooperative’s technological
infrastructure and running costs.
Fund owned by the cooperative: A fund owned by
the cooperative will be created to grant funding from
the cooperative to specific research projects. Thanks to
this mechanism, the cooperative will be able to actively
influence the research agenda and promote research in
areas that are not currently being explored. Therefore,
thanks to this fund, the cooperative is able to contribute
to the research fields that the cooperative considers to
be the most important, not only with data, but also with
financial resources.
Cooperative member wallets: Each cooperative member
will have a personal SalusCoin wallet. Members receive
SalusCoin when: i) they contribute with their data to the
cooperative (more data equals more SalusCoin); ii) they
actively participate in the governance processes of the
cooperative. Citizens can use SalusCoin to acquire services
from accredited service providers, or to fund research
projects promoted by the cooperative.
3.1.5. Relational agreements
The relationships between the parties are bound by
a set of contractual agreements that outline the ethical
behavior, value exchange, technological standards, terms
of data use, etc., to which the parties must agree. Various
types of agreements have been identified. Some of
them apply univocally to all the parties, while others are
specific to certain groups. These agreements require the
management of the divergent interests of the parties, by
addressing issues such as licensing, intellectual property
and data protection. In order to meet the cooperative’s
commitment to transparency, an important future project
would be to explore ways to make these agreements
easily understandable to everyone, in order to allow
cooperative members to make an informed decision.
An important role of the cooperative is to ensure
that these agreements are respected. Legal contracts
should include a clause that states that the cooperative
is not responsible in cases of non-compliance by third
parties.
The specific content of these agreements would be
determined by the collective decision of group members.
Some important aspects to take into account are

40
SALUS.COOP
CHAPTER 3
highlighted below.
ETHICAL AGREEMENT. This agreement is addressed
to all the parties involved in the model. It defines the
cooperative’s aims and the ethical principles that all the
parties agree to abide by. It also states the permitted
purposes of the activities related to data sharing and
use. Two important dimensions of the ethical code are
transparency and participation. All the parties involved
must agree to be open in terms of the processes
that involve data transfer and use. The cooperative is
committed to facilitated participation in all decision-
making processes and offering democratization tools that
allow members to easily participate.
The ethical agreement works as a declaration of
intent and is not therefore legally binding. Parties involved
in the Salus model can however be held to account for
breach of an ethical agreement by cooperative members.
Moreover, some ethical dimensions are included in
the other legally binding contracts signed by Salus
participants.
MEMBERSHIP AGREEMENT. This regulates the rights and
duties of cooperative members and it establishes that
only individuals can become members of the cooperative.
Some of the members’ rights are: the right to vote in
decision-making processes, the right to set (and update)
the terms of use and access to data, the right to be
informed about data usage, and the right to withdraw
at any time. A person who wants to become member of
the cooperative must agree to give permission for the
cooperative to access his/her data, under the conditions
stipulated by the individual, as established by his/her
rights.
DATA TRANSFER AGREEMENT. This regulates the
contractual relationship between data keepers and the
cooperative. The agreement establishes the right of
data keepers to be ‘rewarded’ with data that are better-
structured than the initial data provided. In other words,
data will be returned to them after having been structured
and aggregated. The agreement also establishes that
data will be processed in an anonymous and safe way,
according to existing laws.
SERVICES AGREEMENT. This regulates the contractual
relationship between the service provider companies and
the cooperative, as well as service provider companies
and the third parties (data users, cooperative members,
data keepers) that receive the services. Service provider
companies are not part of the cooperative, and they do
not therefore have voting rights. However, to participate
in the model, companies must receive accreditation issued
by the cooperative. The service agreement could also
concern IP ownership and the licensing of the services
developed by using the cooperative dataset.

41
SALUS.COOP
CHAPTER 3
EXPLOITATION AGREEMENT. This agreement is for data
users who make requests for accessing member data.
The core of this agreement regulates the allowed data
usage, the conditions under which data are donated,
and the obligations to which data users must adhere. It
also covers the criteria that govern the economical flow
between the parties. By establishing the kind of data
usage that is allowed, this agreement plays a critical role
in safeguarding the cooperative’s principle of collective
benefits, ensuring that data are actually used to accelerate
research and generate collective benefits.
This agreement also concerns IP ownership, licensing
and publication conditions, in order to address divergent
interests. In this regard, an important objective of this
agreement is promoting licensing that ensures the lowest
possible drug price (e.g. royalty free, non-exclusive),
limited confidentiality and the open sharing of knowledge
(e.g. creative commons, open-access publications).

42
SALUS.COOP
CHAPTER 3
Data keepers
Public health
centers
Private health
centers
Others
Data users
Universities
Research centers
Clinical research
companies
Others
PREPARATION
ProvidersProviders
EXTRACTION
Identity and security
Registry of agreements
Information catalog
Management System of the Cooperative
GOVERNANCE
Cooperative
FIGURE 8. Technological framework.
3.2 Technological
framework
Four core components define the
technological framework of Salus:
Identity and Security
Management System
This component is made up of
different levels:
• The first would allow members
of the cooperative to be
identified with a unique ID,
which would be related to IDs
from other sources (CIP, Health
insurer services card, etc.). This
would make it possible to access information
about a particular member on all the databases
on which his/her information is stored.
• The second would make it possible to identify
data keepers who have data from members of
the cooperative.
• The third would identify data users who wish
to make use of data under the established
conditions.
• The final level would identify different service
providers, which would contribute to new
services for the other parties participating in the
model.
With regard to the last three points, the type of
identification used to guarantee a person’s identity would
require the implementation of digital certificates.
Security model would affect: (i) the four
components mentioned above, where it is necessary

43
SALUS.COOP
CHAPTER 3
to highlight the need for exhaustive controls over the
traceability of information flows, and (ii) the parties
that can intervene in the Salus model, describing their
commitments in the different smart contracts. A member
of the cooperative would always be able to exercise their
rights, as established in the Data Protection Act (Ley
Orgánica 15/1999, de 13 de diciembre, de Protección de
Datos de carácter personal), with respect to the Access,
Rectification, Cancellation and Opposition (ARCO) of the
information.
In the future access model to be developed, it would
be essential to maintain trust in terms of security and
confidentiality. This is key to ensure an univocal citizen
identification system. The system would include security
requirements and would guarantee the traceability of any
access to the information, which would then be available
for consultation by members of the cooperative whenever
required.
Registry of Agreements
This technological component would guarantee
that all of the agreements between the different parties
in the model (section 3.1.5) would be fulfilled under the
established conditions. To this end, a notary function could
be implemented through Blockchain technology (see next
section).
Information Catalog
The information catalog consists of an index
with statistical information on cooperative members.
This repository should make queries to different data
keepers in order to keep the index up-do-date, for
example via 24/7 web services technology. This index
would make it possible to identify the availability of the
type of information required by data receivers, as well
as the location of the information with the different
data keepers, which would allow queries to respond
to different information requests. This model does not
foresee a centralized information repository, but rather
a distributed repository in which information is stored in
the different data keepers’ servers. A decentralized model
would avoid the risk of a cyber-attack and would establish
a basis of universal trust.
Management System of the Cooperative
The management system would allow the different
operations of the cooperative to be carried out. It would
make it possible, for example, to manage the different
mechanisms of remuneration for all the parties of the
model (section 3.1.4) as well as the other economic
flows for the cooperative. The system could also include
functionalities that support both the management and

44
SALUS.COOP
CHAPTER 3
governance decisions of the cooperative, as well as other
transversal decisions affecting members.
The potential for applying Blockchain
technology in Salus model
The disruptive nature of Blockchain technology can
be seen in the field of finance, but a trend towards new
forms of use is emerging2
. Bitcoin is perhaps the most
widely known use of Blockchain. Health industry giants like
Merck or Phillips have already started venturing into this
technology in search of more efficient solutions for the
sector.
An emerging technology like Blockchain could solve
some of the challenges associated to the Salus model,
especially those related to the transparency and flexibility
of the internal governance processes in the cooperative,
as well as the distributed model of data.
One such future application of this technology could
be in the Registry of Agreements component, wherein a
“notary function” could be implemented. Through the use
of smart contracts, this function would verify and validate
the different types of agreements that are established
between the parties in the cooperative3
. The technical
development of smart contracts is very new. However,
there are already several protocols (e.g. Ethereum) and
2 Donnelly, J., Healthcare: Can the Blockchain optimize and
secure it? BitCoin Magazine, 12 Jan.2016
3 Barbiero A. ¿Podría la tecnología Blockchain ser una alter-
nativa para las apps de salud? Co-Salut 13 Aug. 2015 https://goo.gl/
vAQvyE
platforms, under different levels of development, that are
working towards such contracts becoming an everyday
occurrence.
Identity and Security is another component in
which Blockchain could be used. It is currently difficult
to verify whether a user is who he claims to be, and this
technology shows promise in resolving this kind of issue.
Blockchain technology will allow users to create their own
tamper-proof digital identity, a kind of Blockchain ID that
will soon replace the usernames and passwords that are
used today4
. The public key infrastructure (PKI) works like
a safe deposit box with two keys, one to encode and one
to decode. The cooperative should be safe as long as its
members do not share their PKI with someone else.
Blockchain also offers technological advantages in
terms of distributed architecture, like the one that is
envisioned in the Salus model. Blockchain would allow
data to stay where it has been originated, thus avoiding
the need for a centralized repository.
“For the first time ever, we have a platform
that ensures trust in transactions and much
recorded information no matter how the other party
acts.” Don Tapscott (Donnelly J., Healthcare: Can the
Blockchain Optimize and Secure It? BitCoin Magazine 12
Jan. 2016)
4 Tapscott, D., & Tapscott, A. (2016). Blockchain Revolution:
How the Technology Behind Bitcoin is Changing Money, Business, and
the World. Penguin.

45
SALUS.COOP
CHAPTER 3
Seven Principles of Blockchain
The seven principles of a Blockchain economy, as
proposed by Don and Alex Tapscott1
, are well-aligned
with the principles defined in the Salus model.
Principle 1: Networked Integrity
Due to the fact that the Blockchain is spread out
over thousands of computers, it is “networked” and
participants in a Blockchain economy are incentivized
to maintain “integrity”, as every interaction is indelibly
recorded. Through networked integrity that relies
on a consensus to clear transactions (through so-
called “Proof of Work” or “Proof of Stake”), the use of
intermediaries is no longer required.
Principle 2: Distributed Power
The system distributes power across a peer-to-
peer network with no single point of control. No one
person or organization has a disproportionate control
over the whole, or access to an overly large amount of
data.
Principle 3: Value as Incentive
All stakeholders have aligned incentives. It is
possible to work for your own interests, while also
benefitting society as a whole. Reputation matters.
This is true in the real world as well, but in the
virtual, Blockchain economy, it would likely be less
concentrated on certain established powers.
1 Tapscott, D., & Tapscott, A. (2016). Blockchain
Revolution: How the Technology Behind Bitcoin is Changing
Money, Business, and the World. Penguin.
Principle 4: Security
Given the distributed nature of Blockchain technology,
there is no central point of failure, and if one person
behaves recklessly their capacity to cause damage is
limited.
Principle 5: Privacy
People get to be in complete control of their
data. The Blockchain protocols make it possible to
choose the level of privacy users are comfortable with
in any given transaction or environment. It helps better
manage identities and interaction with the world.
Principle 6: Rights are preserved
Rights and freedoms are clear and enforceable,
as they become part of the Blockchain. Smart contracts
are put in place, which basically allow transactions to
proceed only when predefined benchmarks have been
reached and agreed upon by all parties.
Principle 7: Inclusion
Everyone in the world should have the ability to
participate in the global Blockchain economy. The aim
is to reduce the barriers to participation, and to enable
the greatest number of people and parties to have
technologies and services at their disposal, under the
premise that the economy works best when it works
for all of us.

46
SALUS.COOP
CHAPTER 3
“We will develop a common fabric for the
entire industry. A blockchain to run the entire
healthcare continuum: the free and instant transfer
of healthcare data.” Micah Winkelspecht, Gem Health
(Prisco, G. The Blockchain for Healthcare: Gem Launches
Gem Health Network With Philips Blockchain Lab,
BitCoin Magazine 26 April 2016)
Principal success factors and challenges for the
model
Instrumental factors for the success of the Salus
model and the challenges for the same, in terms of the
technological definition, will be focused on committing to
the information catalogs for the different members of the
cooperative. Moreover, they will guarantee the application
of international standards (technological, semantics, etc.)
that allow the availability of structured information. To
share among the range of different parties and information
sources, it is essential to have univocal identities, not
only for the cooperative members, but also for the
information content (index) in order to avoid misleading
interpretations.
Organizational factors for success and overcoming
challenges will be focused on ensuring transparency and
information security. Advances on these lines are the basis
for promoting trust, and therefore the participation of
citizens in the model.
3.3. Legal framework
To set out the legal guidelines for the project, the
laws that constitute the applicable normative framework
for the cooperative have been identified to ensure the
integrity of the health data and that of the members of
the cooperative.
According to our analysis there does not seem to be
any legal impediment for the deployment of the proposed
model.
Three main categories of regulations have been
identified:
• Laws related to data protection: Law 15/1999,
the Data Protection Act; new regulation of the
European Parliament, EU 2016/679
• Laws related to patients information: Law
41/2002 on Patient Autonomy; Law 21/2000,
on the Rights to Information Concerning
Patient Health and Autonomy and Clinical
Documentation
• Laws related to cooperatives: Catalan Law
12/2015 on Cooperatives
REGULATIONS RELATED TO DATA PROTECTION
Law 15/1999, the Data Protection Act, and
recent European Reform
As a general law, article 7.3 of Law 15/1999 states the
following “personal data that refers to racial origin, health
and sexual relations may only be collected, processed
and transferred when, in the general interest, a law or the

47
SALUS.COOP
CHAPTER 3
affected person expressly provides consent”. This article
determines the fundamental right to the protection of
personal data, and expresses the ability of the citizen to
cede information to an interested party.
This Law also regulates the protection of health
data, and includes recognition of the rights of access,
rectification, cancellation and opposition that make up the
system of guarantees, and form an essential aspect of the
right to personal data protection.
The new European Data Protection Regulation
(EU 2016/679)
The new European Data Protection Regulation also
applies on similar lines. This reform aims to give citizens
back the control of their personal data, and to ensure
high protection standards in the digital environment
throughout the EU. Among other provisions, the new law
also includes:
• The right to “forgetfulness”, through the
correction or suppression of personal data,
• The need for a “clear and affirmative consent”
to the processing of their personal data by the
person concerned,
• “Portability”, or the right to transfer data to
another service provider
The different member states of the EU will have a
period of two years to apply the changes in the directive
to national legislation.
REGULATIONS RELATED TO PATIENT DATA
Law 21/2000, on the Rights to Information
Concerning Patient Health and Autonomy and
Clinical Documentation.
Article 3.1 of this Law establishes that “The holder of the
right to information is the patient. Persons linked to him
or her must be informed to the extent that is expressly or
tacitly permitted.”
With regard to the right to access medical records
(documents accrediting a healthcare relationship), Article
13.3 states: “The right of the patient to access medical
records may also be exercised by a representative,
provided that it is duly accredited.”
Both provisions ensure legal representation by the
cooperative, with express authorization by the citizen,
for access to clinical information. This complies with the
conditional donation of data by citizens in accordance
with the ethical principles of the cooperative.
Law 41/2002, on Patient Autonomy
Article 5.1 of this law tacitly establishes that “The patient
holds the right of information. Those related to the
patient, through family or by law, will also be informed to
the extent that the patient expressly or tacitly allows it.”
Regarding the right of access to clinical information,
this law establishes the following in article 18.1 “The patient
has the right of access [...] clinical history documentation
and to obtain a copy of the data contained therein”, as

48
SALUS.COOP
CHAPTER 3
well as, “The right of the patient to access medical records
can also be exercised by a duly accredited representative”.
REGULATIONS RELATED TO COOPERATIVES
Law 12/2005 of Catalan Cooperatives
The Law is based on the historical general principles of
the International Co-operative Alliance (ICA) and, on
an international consensus that a cooperative is, “A co-
operative is an autonomous association of persons united
voluntarily to meet their common economic, social, and
cultural needs and aspirations through a jointly-owned
and democratically-controlled enterprise.” Likewise, it
covers the principles that should govern the activity
of cooperatives in Catalonia, which are included in
the present law, and are those defined by the ICA. The
principles are the following:
• Voluntary and open membership;
• Democratic member control;
• Member economic participation;
• Autonomy and independence;
• Education, training and information;
• Co-operation among co-operatives,
• Concern for community.
The cooperative principles covered in this law will
shape the management and principles of the cooperative
in the Salus model.
Other regulation related to the Salus
model
Ley 16/2003, de 28 de mayo, de cohesión
y calidad del Sistema Nacional de Salud
(Law 16/2003, of 28 May, on the cohesion and
quality of the National Health System):
Citizens receiving the benefits of the National
Health System will have the right to healthcare,
health information and documentation, in
accordance with Law 41/2002, of 14 November
(Article 7.2).
Ley 44/2003, de 21 de noviembre, de
ordenación de las profesiones sanitarias
(Law 44/2003, of 21 November, on the organization
of health professions):
Patients have the right to receive information
according to what is established in Law 41/2002
(article 5.1.f)
Ley 55/2003, de 16 de diciembre, del
Estatuto Marco del personal estatutario
de los servicios de salud (Law 55/2003, of
16 December, on the Statutory Framework of the
Statutory Staff in Health Services):
The duties of statutory staff include the duty to
duly inform users and patients about their care
process, as well as the services available (article
19.h.), in accordance with the rules and procedures
applicable to each case and within the scope of
their responsibilities.

50
SALUS.COOP
CHAPTER 4
CONCLUSION
Conclusion
Access to large sets of health data is key to advancing
medical research. However, today, health data are
fragmented and guarded in silos, thus hindering their access
and sharing by researchers.
Diverse initiatives around the world, ranging from
public to private organizations, have attempted to
overcome these hindrances by proposing new ways to share
and generate health data. However, only a few of them
legitimize data ownership for citizens and enable them to
exercise control over the fate of their data.
The investigation and proposed frameworks
presented in this report aimed to explore how a citizen-
driven model of collaborative governance and management
of health data, which enables citizens to collectively share
data, can accelerate research and innovation in healthcare.
Towards this end, Ideas for Change conducted field and
desk research aimed to gather opinions and suggestions
from key agents in the healthcare sector, and analyze the
current health data ecosystem along with a sample of
existing initiatives.
The findings of this research highlighted four
fundamental principles that should be considered the pillars
of any citizen-driven governance model for health data
management:
(i) citizens should be given the right to decide under which
conditions they want to donate their health data;
(ii) the use of data by any parties involved should provide
a clear benefit to society (e.g. the resulting research outputs
are made universally available under an open license);
(iii) incentives (non-monetary) should be offered to
individuals to motivate them to donate their data;
(iv) a participatory governance model should be deployed
to guarantee the collective benefits of data use and citizen
control over the fate of their data.
Drawing upon the results of this research, a
cooperative governance model that enacts these four
principles has been designed. The proposed model considers
the creation of an ecosystem of actors that provide
different types of value to the model: data, services, and
economic. The relationships among the agents are regulated
by a set of contractual agreements that ensure compliance
with the principles established by the cooperative members.
Openness and transparency are fundamental to the model,
which enables data contributors to participate in decision
making. The investigation and resulting approach presented
in this report offer a new vision to reshape the health data
sharing ecosystem. We trust that, if fully applied, this vision
has the potential to deliver systemic change.
Data: scarcity abundance
Management: individual collective
Channels: intermediaries direct
Knowledge: asymmetry symmetry of information
Publications: selective integral
Actors: a certain number multiplicity
Innovation: on products on processes

Salus.Coop Informe Final

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Salus.Coop Informe Final

Similar to Salus.Coop Informe Final (20)

More from Andrea Barbiero

More from Andrea Barbiero (7)

Recently uploaded

Recently uploaded (20)

Salus.Coop Informe Final